Table of contents

Characteristics of MHD instabilities and their impacts on plasma confinement are studied in current free plasmas of the Large Helical Device. Spontaneous L–H transition is often observed in high beta plasmas close to 2% at low toroidal fields (B_t ⩽ 0.75 T). The stored energy starts to rise rapidly just after the transition accompanying the clear rise in the electron density but quickly saturates due to the growth of the m = 2/n = 3 mode (m and n: poloidal and toroidal mode numbers), the rational surface of which is located in the edge barrier region, and edge localized mode (ELM) like activities having fairly small amplitude but high repetition frequency. Even in low beta plasmas without L–H transitions, ELM-like activities are sometimes induced in high performance plasmas with a steep edge pressure gradient and transiently reduce the stored energy up to 10%. Energetic ion driven MHD modes such as Alfvén eigenmodes (AEs) are studied in a very wide range of characteristic parameters (the averaged beta of energetic ions, ⟨β_b∥⟩, and the ratio of energetic ion velocity to the Alfvén velocity, V_b∥/V_A), of which range includes all tokamak data. In addition to the observation of toroidicity induced AEs (TAEs), coherent magnetic fluctuations of helicity induced AEs (HAEs) have been detected for the first time in NBI heated plasmas. The transition of a core-localized TAE to a global AE (GAE) is also observed in a discharge with temporal evolution of the rotational transform profile, having a similarity to the phenomenon observed in a reversed shear tokamak. At low magnetic fields, bursting TAEs transiently induce a significant loss of energetic ions, up to 40% of injected beams, but on the other hand play an important role in triggering the formation of transport barriers in the core and edge regions.

Pellet injection gives rise to a rich variety of observations on m = 1 island dynamics. Pellet ablation near the q = 1 surface triggers fast growth of an island; as the island reaches a large amplitude, magnetic reconnection mixes the plasma centre with the q = 1 pellet fuelled region, so enhancing the effective pellet deposition depth. Long-lived m = 1 islands that coexist with sawtooth crashes are accompanied by higher harmonics (m/n = 1 with m > 1, where m is the poloidal mode number and n is the toroidal one) near the q = 1 surface; progressive phase shifts observed in the phase of such harmonics indicate island deformation by sheared viscous flow. Coupling between m = 1 and m = 2 islands is observed in sawtooth-free plasmas. Reconnection at the q = 2 surface is driven by a toroidal sideband of the m = 1 island; coupling is associated with strong braking of the m = 1 island.

Two approaches are examined for stabilization of the resistive wall mode (RWM) of toroidal mode number n = 1 in an advanced ITER scenario. Active feedback control, with the present coil design and poloidal sensors placed just inside the inner wall, can be very efficient in stabilizing the RWM. Within the voltage limit of the present design for the feedback coils and conservative constraints on performance, the plasma pressure can be increased up to at least 70% between the no-wall and ideal-wall beta limits. Stabilization of the RWM by toroidal plasma rotation depends on the rotation profile as well as on the model for ion Landau damping. Feedback control of rotating plasmas for the advanced scenario is considered. The effect of the blanket is also studied using a simplified model.

The first suppression of the important and deleterious m = 2/n = 1 neoclassical tearing mode (NTM) is reported using electron cyclotron current drive (ECCD) to replace the 'missing' bootstrap current in the island O-point. Experiments on the DIII-D tokamak verify that the maximum shrinkage of the m = 2/n = 1 island occurs when the ECCD location coincides with the q = 2 surface. The DIII-D plasma control system is put into a 'search and suppress' mode to make small changes in the toroidal field to find and lock onto the optimum position, based on real time measurements of dB_θ/dt, for complete m = 2/n = 1 NTM suppression by ECCD. The requirements of the ECCD for complete island suppression are well modelled by the modified Rutherford equation for the DIII-D plasma conditions.

Modelling of carbon pellet injection into hot hydrogen plasma is performed. Models of ionization and radiation processes previously used for descriptions of the ablation clouds are discussed. It is demonstrated that the assumption made by other authors that the plasma is transparent significantly overestimated ionization lengths for carbon. The optical thickness of the cloud is estimated. With typical ablation rates, the cloud in the vicinity of the pellet is opaque for resonant radiation in lines, in contrast to the visual transparency for non-resonant photons. The ionization by incident electrons is not sufficient to provide the short experimental cloud sizes. A reduced model for the ion stripping dynamics and energy losses in optically thick plasmas is proposed. Two very important phenomena are taken into account: partial trapping of resonant photons in the cloud and ionization from the excited state. The model is included into the MHD pellet code LLP. Calculations with the new model show at least a qualitative agreement with the experimental sizes of the carbon clouds, in contrast to the models used earlier.

Impurity behaviour in JET internal transport barrier (ITB) discharges with reversed shear has been investigated. Metallic impurities accumulate in cases with too strong peaking of the main ion density profile. The accumulation is due to inwardly directed drift velocities inside the ITB radius. The strength of the impurity peaking increases with the impurity charge and is low for the low-Z elements C and Ne. Transport calculations show that the observed behaviour is consistent with dominant neoclassical impurity transport inside the ITB. In some cases, MHD events in the core flatten the radial profile of the metallic impurity.

In this study, the fixed boundary plasma MHD equilibrium problem is solved by the finite element method; then, by perturbing the flux at the plasma boundary nodes, linear formulae are derived linking the variation of several plasma parameters of interest to the variation of the currents flowing in the external circuits. On the basis of these formulae it is shown how it is possible to efficiently solve two central problems in plasma engineering, namely (1) the optimization of the currents in a given set of coils necessary to maintain a specified equilibrium configuration and (2) the derivation of a linear dynamic model describing the plasma axisymmetric displacement (n = 0 mode) about a given magnetic configuration. A case study—based on the ITER reference equilibrium magnetic configuration at burn—is analysed both in terms of equilibrium currents optimality as well as axisymmetric stability features. The results obtained by these formulae are also compared with the predictions of a non-linear free boundary code and of a linear, dynamic model. As shown, the formulae derived here are in good agreement with such predictions, confirming the validity of the present approach.

The H-1NF heliac is a medium-sized heliac stellarator experiment with major radius R = 1 m, and average plasma minor radius a = 0.15–0.2 m. Its 'flexible-heliac' coil set permits precise variation in the value and shape of the rotational transform profile, with regions of both positive and negative shear. Operation at low fields (B < 0.2 T) with argon plasmas heated by helicon waves produces plasmas that have large ion Larmor radii (ρ_i/a ∼ 0.4) and show confinement transitions at low power like those in the edge of large devices, yielding fundamental measurements concerning electric fields and zonal flows. At a higher field (0.5 T), precise rotational transform scans with H–He plasmas heated by ICRF show resonant equilibrium and stability phenomena which depend on the value of the rotational transform at the radius of zero shear.

The results of the m = 2, n = 1 tearing mode natural frequency measurements in the course of the mode-locking by a controlled halo-current are presented. The halo-current in the rail limiter–plasma–vacuum vessel–external circuit with a controlled EMF source–rail limiter loop was used. The halo-current represented a superposition of a direct-current pulse for the mode-locking and an oscillating-current component for the natural frequency measurements by the resonant response of the mode. In the experiment, the mode signal oscillations at the frequency of the halo-current oscillating component were observed even if the mode rotation stopped under the pulse of the halo-current. The maximum amplitude of these mode oscillations was achieved when the frequency of the halo-current oscillating component was equal to the frequency of the mode before switching on the halo-current. It means that under the conditions of the T-10 experiment, there was no variation of the mode natural frequency upon termination of the mode rotation by the halo-current. This mode behaviour can be attributed to the well-known effect of frequency locking in an externally driven non-linear oscillator. The experiment was simulated with a model of a rotating non-linear tearing mode in the presence of the halo-current.

The structure of the radial electric field and heat transport at the magnetic island in the large helical device (LHD) are investigated by measuring the radial profile of the poloidal flow with charge exchange spectroscopy and measuring the time evolution of the electron temperature with ECE. A vortex-like plasma flow along the magnetic flux surface inside the magnetic island is observed when the n/m = 1/1 external perturbation field becomes large enough to increase the magnetic island width above a critical range (15–20% of minor radius) in LHD. This convective poloidal flow results in a non-flat space potential inside the magnetic island. The sign of the curvature of the space potential (∂²Φ/∂r², where Φ is the space potential) depends on the radial electric field at the boundary of the magnetic island. The heat transport inside the magnetic island is studied with a cold pulse propagation technique. The experimental results show the existence of radial electric field shear at the boundary of the magnetic island and a reduction in heat transport at the boundary and inside the magnetic island.

The 'fishbone' fast wave travelling wave antenna was developed for LHD to provide a capability for rotational transform profile control by current drive. The fishbone antenna is equivalent to two combline antennae stacked vertically. The antenna operates around 75 MHz and excites a wavenumber of 14 m⁻¹ when the phase difference between adjacent current straps is 90°. A test of a combline antenna with plasma load on the TST-2 spherical tokamak suggested the possibility that this type of antenna does not need to be installed in the immediate vicinity of the last closed flux surface. Optimization of the design was performed based on measurements on mock-up antennas and model calculations. In the fishbone antenna, controlled excitation of the even mode (with currents in the top and bottom halves of a current strap in the same direction) is necessary. A predominant excitation of the even mode was realized in the LHD fishbone antenna with simulated loading by selecting an appropriate operating frequency.

The modifications in plasma parameters and heating power under conditions of pulsed poloidal current drive (PPCD) in Reversed Field eXperiment (RFX) are modelled by the one-dimensional transport code RITM (Radiation from Impurities in Transport Model). Several channels for the PPCD effects on plasma confinement are taken into account: dissipation of externally driven poloidal currents, reduction in the electric field induced by the dynamo and decrease in magnetic fluctuations. It is demonstrated that the external currents are dissipated at the plasma edge and they cannot, alone, lead to any change in the temperature profile. The reduction in the dynamo electric field results in central peaking of the current and Ohmic power densities, accompanied, however, by a very modest rise in the central temperature. By taking into account the decrease in the level of magnetic fluctuations (predicted for PPCD conditions by non-linear MHD simulations) and the corresponding reduction of transport in the stochastic magnetic field, RITM reproduces, in agreement with experiments, a significant increase in the central temperature and a reduction in the loop voltage.

The main objective of this paper is investigation of methods for reduction of divertor heat loads in order to increase the lifetime of divertor tiles in future fusion reactors. Special emphasis is given to studies of reduction of transient heat loads due to edge localized modes (ELMs). Two methods are compared: argon seeded type-I ELMy H-modes and nitrogen seeded type-III ELMy H-modes. In both scenarios, the impurity seeding leads to a reduction in the pedestal energy and hence a reduction in the energy released by the ELM. This consequentially reduces the power load to the divertor targets. At high radiative power fractions in type-III ELMy H-modes, part of that released ELM energy (small ELMs, below 20 kJ) is dissipated by radiation in the scrape off layer (SOL). Modelling of the ELM mitigation supports the experimental findings. This ELM mitigation by radiative dissipation is not effective for larger ELMs. In between ELMs, the plasma is detached and radiates strongly from the X-point region. During an ELM, the nitrogen in the X-point and divertor region becomes ionized into more weakly radiating higher charge states and the plasma re-attaches for large ELMs. At JET, argon radiates predominantly in the main plasma and not so much in the cold divertor region. Hence, the effect of radiative dissipation of ELM heat fluxes by argon is very low due to the limited argon density in the divertor region. Nevertheless, both scenarios might be compatible with an integrated ITER scenario, with respect to acceptable divertor lifetime and acceptable confinement.

We review and test the peeling–ballooning model for edge localized modes (ELMs) and pedestal constraints, a model based upon theoretical analysis of magnetohydrodynamic (MHD) instabilities that can limit the pedestal height and drive ELMs. A highly efficient MHD stability code, ELITE, is used to calculate quantitative stability constraints on the pedestal, including constraints on the pedestal height. Because of the impact of collisionality on the bootstrap current, these pedestal constraints are dependent on the density and temperature separately, rather than simply on the pressure. ELITE stability calculations are directly compared with experimental data for a series of plasmas in which the density is varied and ELM characteristics change. In addition, a technique is developed whereby peeling–ballooning pedestal constraints are calculated as a function of key equilibrium parameters via ELITE calculations using series of model equilibria. This technique is used to successfully compare the expected pedestal height as a function of density, triangularity and plasma current with experimental data. Furthermore, the technique can be applied for parameter ranges beyond the purview of present experiments, and we present a brief projection of peeling–ballooning pedestal constraints for burning plasma tokamak designs.

Modification of JT-60 as a full superconducting tokamak (JT-60SC) is planned. The objectives of the JT-60SC programme are to establish scientific and technological bases for steady-state operation of high performance plasmas and utilization of reduced-activation materials in an economically and environmentally attractive DEMO reactor. Advanced fusion technologies relevant to the DEMO reactor have been developed for the superconducting magnet technology and plasma facing components of the JT-60SC design. To achieve a high current density in a superconducting strand, Nb₃Al strands with a high copper ratio of 4 have been newly developed for the toroidal field coils (TFCs) of JT-60SC. The R&D to demonstrate the applicability of the Nb₃Al conductor to TFCs by a react-and-wind technique has been carried out using a full-size Nb₃Al conductor. A full-size NbTi conductor with low ac loss using Ni-coated strands has been successfully developed. A forced cooling divertor component with high heat transfer using screw tubes has been developed for the first time. The heat removal performance of the carbon fibre composite target was successfully demonstrated on an electron beam irradiation stand.

Plasma parameters, like the internal inductance, l_i, and the diamagnetic poloidal beta, β_DIA, are of particular relevance for a reliable real-time control system of next step tokamaks. These and other quantities have been obtained at Joint European Torus (JET) with a method that uses the Shafranov integrals S₁, S₂ and S₃. Indeed they allow the direct calculation of the Shafranov parameter Λ = β_MHD + l_i/2 and of β_DIA. Moreover, in discharges with a sufficiently high elongation (k > 1.3 typically), the internal inductance can be separated from the MHD poloidal beta, β_MHD, and calculated independently, through the Shafranov integrals, with a precision that is more than satisfactory for real-time applications. It is worth mentioning that, since S₁, S₂ and S₃ are integrals defined on the plasma boundary, a specific algorithm, depending on the fast code XLOC, has been expressively developed to determine this quantity.

The method to determine the plasma parameters has been verified off-line, benchmarking its outputs with the estimates of the most sophisticated equilibrium codes available. The results of this systematic comparison have been very encouraging both in the limiter and x-point phases of the discharges and on all the investigated plasma configurations. The computational time, necessary to determine the plasma boundary and about 50 signals is only about 1.5 ms on a PC equipped with a 400 MHz Pentium II, well below the 10 ms constraint of JET real-time applications. The code has therefore been implemented on-line, and its outputs have already been exploited to achieve the feedback control of some plasma parameters.

We report recent progress in the study of neoclassical internal transport barriers (N-ITBs) in the compact helical system (CHS) experiment. N-ITBs have been observed in electron cyclotron (EC) heated NBI plasmas. The improved confinement region for electrons is expanded compared with that of previous N-ITB plasmas heated by only EC heating. Moreover, the ion temperature is found to be increased simultaneously by two to three times (T_i(0) ∼ 400–500 eV), with a steep gradient region at ρ ∼ 0.6. From measurement of the averaged peak energy intensity using a soft x-ray CCD camera, it is confirmed that the impurity confinement is also improved with N-ITBs. The radial electric field is observed to bifurcate into ion and electron root (E_r ∼ 15 kV m⁻¹), and a rather large electric field shear (dE_r/dr ∼ −300 kV m⁻²) is produced in the layer between ion and electron root.

Recent performance analysis has improved confidence in achieving Q (= fusion power/auxiliary heating power)⩾ 10 in inductive operation in ITER. Performance analysis based on empirical scalings shows the feasibility of achieving Q ⩾ 10 in inductive operation, particularly with improved modelling of helium exhaust. Analysis has also indicated the possibility that ITER can potentially demonstrate Q ∼ 50, enabling studies of self-heated plasmas. Theory-based core modelling indicates the need for a high pedestal temperature (3.2–5.3 keV) to achieve Q ⩾ 10, which is in the range of projections with presently available pedestal scalings. Pellet injection from the high-field side would be useful in enhancing Q and reducing edge localized mode (ELM) heat load in high plasma current operation. If the ELM heat load is not acceptable, it could be made tolerable by further tilting the target plate. Steady state operation scenarios at Q = 5 have been developed with modest requirements on confinement improvement and beta (H_H98(y,2) ⩾ 1.3 and β_N ⩾ 2.6). Stabilization of the resistive wall modes (RWMs), required in such regimes, is feasible with the present saddle coils and power supplies with double-wall structures taken into account. Recent analysis shows a potential of high power steady state operation with a fusion power of 0.7 GW at Q ∼ 8. Achievement of the required β_N ∼ 3.6 by RWM stabilization is a possibility. Further analysis is also needed on reduction of the divertor target heat load.

Four types of plasma current ramp-up experiments in full non-inductively lower hybrid current driven (LHCD) plasmas were executed in TRIAM-1M: (1) current start-up by a combination of electron cyclotron resonance heating (ECRH) and LHCD, (2) tail heating by additional LHCD, (3) bulk heating by ECRH and (4) spontaneous ramp-up by a transition to enhanced current drive (ECD) mode. The time evolutions of plasma current during four types of ramp-up phase were adjusted by a simple model with two different time constants, which are a time defined by the total current diffusion time and a time constant for improving the current drive efficiency. In the case of (1) and (4), the latter time constant is significant during the current ramp-up phase. The improvement in the current drive efficiency in the ECD mode is likely to be caused by the increase in the effective refractive index along the magnetic field of the lower hybrid wave.

Critical stability issues involving magnetic reconnection that are likely to influence the successful operation of burning plasma experiments are addressed. In particular, we discuss: (1) the stabilization of drift-tearing modes in weakly collisional regimes, with particular reference to nonlinear stability, including neo-classical effects; (2) the present understanding of sawtooth oscillations, in particular the interpretation of peculiar sawtooth features in FTU experiments with pellet injection; (3) reconnection near the X-points of magnetic separatrices.

Recent experimental results from the HL-1M tokamak and progress in the HL-2A project are presented in this paper. In HL-1M, strong fishbone instability was observed during off-axis electron cyclotron resonance heating (ECRH). This is the first observation of fishbone instability purely driven by energetic electrons produced by ECRH. The molecular beam injection (MBI) was first proposed and demonstrated in HL-1M. Recently, new results of the MBI experiment were obtained by increasing the pressure of the gas. A stair-shaped density increment was obtained with high-pressure multi-pulse MBI just like the density evolution behaviour during multi-pellet injection. It is shown that injected particles penetrated into the core region of the plasma. HL-2A is a divertor tokamak constructed at SWIP based on the original ASDEX main components. The mission of the HL-2A project is to explore the physics issues involved in an advanced tokamak. For the first phase, divertor (edge plasma) and confinement research will be emphasized. The major parameters of HL-2A are R = 1.65 m, a = 0.4 m, B_t = 2.8 T, I_P = 0.48 MA. The main parameters and characteristics of the subsystems such as power supply, pumping, diagnostics and auxiliary heating are presented in this paper. The first plasma of HL-2A was obtained at the end of 2002.